Extracellular vesicles from Lacticaseibacillus paracasei PC-H1 inhibit HIF-1α-mediated glycolysis of colon cancer

Colorectal cancer is the fourth-leading cause of cancer death worldwide [1]. The factors affecting the incidence of colon carcinoma include diet, environment and intestinal homeostasis [2]. Its treatment methods mainly include surgery and chemotherapy [3]. However, challenges such as chemotherapy resistance [4], poor prognosis [5] and glycolytic state [6] are yet to be addressed. Therefore, we should act quickly to develop a new strategy for combating this disease.

The effects of probiotics in colorectal cancer have been studied extensively. Lacticaseibacillus paracasei is a major probiotic that secretes metabolites [7], exopolysaccharides [8], whole peptidoglycan extracts [9], cell wall protein fractions [10] and extracellular vesicles (EVs) [11] with antitumor activity in colon cancer. EVs are lipid bilayer nanoparticles that can be released by cells from all living kingdoms [12]. Bacterial-derived EVs can carry information from their parent cells and play a pivotal role in bacterial survival, nutrient sensing and participation in bacterial–host communication [13].

The beneficial effects of EVs derived from probiotics have been reported in previous research [14]. EVs produced by Propionibacterium freudenreichii CIRM-BIA 129 displayed an anti-inflammatory effect by modulating the NF-κB pathway [15]. Lactobacillus plantarum Q7-derived extracellular vesicles alleviate dextran sodium sulfate-induced ulcerative colitis by improving the dysregulation of gut microbiota [16]. EVs released by Lacticaseibacillus paracasei inhibited lipopolysaccharide-induced inflammation by augmenting endoplasmic reticulum stress pathway [17]. Moreover, extracellular vesicles released by Lactobacillus rhamnosus GG can inhibit the growth of hepatic cancer cells [18].

HIF-1α, a primary metabolic sensor, can switch the tumor from the aerobic metabolic pathway to the glycolytic pathway [19]. HIF-1α increases the entry of glucose into tumor cells to meet their energy needs by inducing the expression of GLUT1 and enzymes involved in the glycolytic pathway [20]. Even though it has been verified that the activity of glycolysis is increased in colon cancer cells, the effect of Lacticaseibacillus paracasei EVs (LpEVs) on the metabolic activities of colon cancer cells has not yet reported. In this study, the authors examined the impact of Lacticaseibacillus paracasei PC-H1-derived extracellular vesicles on colon cancer cells' metabolic activity.

The results indicate that Lacticaseibacillus paracasei PC-H1 extracellular vesicles suppress colorectal cancer growth by lowering HIF-1-mediated glycolysis. It is believed that these findings provide a theoretical foundation for developing cutting-edge therapeutic approaches for colorectal cancer.

Materials & methods

Bacterial cultures & LpEV isolation

Probiotic Lacticaseibacillus paracasei PC-H1 CGMCC22285 was isolated from healthy human feces samples in northeast China. The isolated bacteria strain was cultivated overnight at 37 °C in De Man, Rogosa and Sharpe medium. To remove large particles or debris from the cultures and to get the bacteria-free culture supernatants, cell-free culture supernatant was collected and then high-speed centrifugation (120,000× g) was used to obtain LpEVs. A nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, MA, USA) was used to evaluate the concentration of purified LpEV. The detailed procedures are provided in the authors' previous paper [11].

Transmission electron microscopy & nanoparticle tracking analysis

Morphology of the LpEVs was determined using transmission electron microscopy (Hitachi, Tokyo, Japan). The size distribution curves and concentration of LpEVs were detected by nanoparticle tracking analysis of ZetaView (Particle Metrix, Meerbusch, Germany).

Cell culture

Human HCT116 colorectal cancer cells were cultured in Dulbecco’s modified Eagle medium with 10% fetal bovine serum and incubated at 37 °C, 5% CO₂.

Labeling of LpEVs

The internalization of LpEVs by HCT116 cells was examined by labeling the membrane and content mRNAs of LpEVs with a ExoGlow™-Membrane EV Labeling Kit (System Biosciences, CA, USA) and ExoGlow™-RNA EV Labeling Kit (System Biosciences, CA, USA), respectively. Then LpEVs were added to the labeling reaction solution after the reaction buffer and labeling dye were combined. The samples were mixed well and incubated in the dark for 30 min. The unlabeled probes were removed by PD SpinTrap G-25 buffer (GE Healthcare Life Sciences, MA, USA). Labeled LpEVs were incubated with cells for 24 h, then observed using confocal microscopy (Yokogawa, Japan) [11].

Cell proliferation assay

Cells (1 × 10⁴ cells/well) were seeded in 96-well plates and then incubated at 37 °C. When the cells were 90% confluent, they were exposed to LpEVs at concentrations of 50 μg/ml, 100 μg/ml, 150 μg/ml and 200 μg/ml over 48 h at 37 °C. WST-1 reagent (Beyotime, China) was added to the wells and the cells were incubated for 1 h at 37 °C. A microplate reader was used to measure the absorbance values at 450 nm.

RNA sequencing & bioinformatic analysis

In order to evaluate the effect of LpEVs on the mRNA levels of HCT116 cells, RNA was collected from the cells that had been exposed to LpEVs at a concentration of 200 μg/ml for 48 h The sequencing libraries were created at Novogene Technology Corporation (Beijing, China) using the NEBNext^® UltraTM Directional RNA Library Prep Kit for Illumina^® (NE, USA). GOseq R package was used to do the Gene Ontology analysis, and p-values less than 0.05 were considered as significantly enriched by differentially expressed genes. Kyoto Encyclopedia of Genes and Genomes enrichment was carried out by KOBAS software. The protein–protein interaction (PPI) study of differentially expressed genes was based on the Search Tool for the Retrieval of Interacting Genes/Proteins database, which contained known and predicted PPIs.

Quantitative real-time polymerase chain reaction

The total RNA was subjected to isolate from each sample using TRIzol reagent. Separated RNA was reverse-transcribed into cDNA using the PrimeScript™ RT Master Mix (Takara, Japan). Then quantitative real-time polymerase chain reaction reactions were carried out using the SYBR Premix Ex Taq™ (Takara, Japan) on a LightCycler 480II Real-Time PCR System (Roche, Switzerland). The relative levels of each gene's expression between groups were calculated using the 2^-ΔΔCT method. Using the 18S gene as the internal control, the mRNA levels of the target genes were standardized. The specific primer sequences are displayed in Table 1.

Western blot analysis

After 48 h of treatment, the total protein of the cells was extracted using radioimmunoprecipitation assay lysis buffer (Beyotime, China) for 30 min on ice.The protein content of each sample was determined using a protein concentration assay kit (Beyotime). SDS-PAGE was used to separate the protein. The HIF-1α (ab51608,1:1000, Abcam, MA, USA), GLUT1 (ab115730,1:1000, Abcam), LDHA (#3582, 1:1000, Cell Signaling Technology, MA, USA) and glyceraldehyde-3-phosphate dehydrogenase (#5174, 1:5000, Cell Signaling Technology) primary antibodies and secondary antibody (A0208, 1:1000, Beyotime) were incubated with the protein. The protein was visualized with an electrochemiluminescence kit (Alphabio, China).

Glucose, lactate & lactate dehydrogenase measurement

HCT116 cells (1 × 10⁴ cells/well) were cultured in 96-well plates and then incubated at 37 °C. Following attaching on plate, the cells were treated with LpEVs at a concentration of 200 μg/ml for 48 h. The glucose consumption, lactate production and lactate dehydrogenase (LDH) levels were measured using a glucose assay kit (ZCIBIO, China), lactic acid assay kit (Jiancheng Bioengineering Institute, China) and the LDH assay kit (Alphabio, China) according to the manufacturers' instructions.

Xenograft experiments

Female athymic BALB/c nude mice were obtained from Charles Fiver (Beijing, China). The mice were kept in a pathogen-free environment with unrestricted access to food and drink. The BALB/c nude mice were randomized into two groups: the control group and the LpEVs group. Subsequently, two groups of mice were subcutaneously injected with 200 μl phosphate-buffered saline containing HCT116 cells (1.5 × 10⁶ cells) or a mixture of HCT116 cells (1.5 × 10⁶ cells) and LpEVs (200 μg/ml). After 30 days, the mice were sacrificed by cervical dislocation under anesthesia. Next, tumors were removed for further analysis. The Institutional Animal Care and Use Committee of Harbin Medical University granted the animal study approval (HMUIRB20210007).

Immunohistochemistry

The detailed immunohistochemistry assay procedures are provided in the authors' previous paper [11]. The primary antibodies are listed in Supplementary Table 1.

Statistical methods

All experiments were repeated more than three times, with their outcomes presented as mean ± standard deviation. A t-test was utilized to analyze differences between the control and LpEV-treated groups, where the value of p < 0.05 was considered statistically significant. GraphPad Prism 5 was used to create the graphs.

Results

Characterization of LpEVs & internalization of LpEVs by HCT116 cells

LpEVs were obtained from the supernatant of ultracentrifuged Lacticaseibacillus paracasei PC-H1. The isolated particles were characterized by electron microscopy magnified 100,000-times. The micrograph revealed that the purified LpEVs were round in shape and had a vesicle-like structure of lipid bilayers (Figure 1A). Particle numbers of each gradient fraction were also detected using nanoparticle tracer analysis. The LpEV particle diameters were mostly concentrated in the 200 nm range (Figure 1B). This finding is consistent with the typical results observed under transmission electron microscopy. Because of its extremely specific membrane sensor, the ExoGlow™ kit can label EVs for fluorescence detection. Entry of many labeled LpEVs into HCT116 cells was observed under the confocal microscopy (Figure 1C). ExoGlow™-RNA EV Labeling Kit is the latest generation of fluorescent labeling reagents that specifically enter EVs and tightly bind mRNAs for fluorescence-based detection. In fluorescence images, many labeled LpEVs-mRNAs were observed in HCT116 cells (Figure 1D).

LpEVs inhibit the proliferation of HCT116 cells

To determine whether LpEVs inhibited the proliferation of HCT116 cells, different doses of LpEVs were cocultured with HCT116 cells for 48 h. In comparison with the control group, LpEVs at the highest concentration tested (200 μg/ml) significantly reduced the proliferation of HCT116 cells (Figure 2).

Identification of differentially expressed genes & functional annotation of HCT116 cells treated with LpEVs

According to high-throughput sequencing, gene expression was altered after LpEVs were administered to HCT116 cells at a high concentration (200 μg/ml). The Gene Ontology enrichment results showed that differentially expressed genes were enriched in glycolytic process (Figure 3A). The differentially expressed genes were then blasted against the Kyoto Encyclopedia of Genes and Genomes database, and the results (Figure 3B) showed that differentially expressed mRNA participates in the HIF-1 signaling pathway, in which GLUT1 was related with the HIF-1α pathway (Figure 3C) and high-throughput sequencing results showed reduced GLUT1 expression. To further explore the relationship between the mRNAs of these differentially expressed genes, the PPI network was constructed using the STRING online database. Highly correlated module analysis showed that LpEVs had an impact on important metabolic enzymes in the glycolytic pathway (Figure 3D).

Expression of HIF-1α, GLUT1 & important metabolic enzymes in colorectal cancer

In order to evaluate the expression of important metabolic enzyme genes in HCT116 cells, highly correlated modules in the PPI network were selected for quantitative polymerase chain reaction detection. In HCT116 cells treated with LpEVs, the mRNA expression levels of LDHA, HK2, ENO2, PGK1, PGAM1 and GP1 were significantly reduced (Figure 4A–F). This outcome is in line with the sequencing results. In RNA sequencing, differentially expressed genes were enriched in the HIF-1 signaling pathway, in which the expression levels of LDHA and GLUT1 were significantly reduced. Both LDHA and GLUT1 are target genes of HIF-1α, and HIF-1α can induce the expression of glycolysis-related genes and enhance the uptake of glucose by cancer cells [21]. Therefore, the authors inferred that LpEVs might inhibit HIF-1α in HCT116 cells to reduce the expression level of GLUT1 or LDHA, which weakens the glycolysis of colon cancer cells and indirectly slows tumor growth. After treating HCT116 cells with LpEVs, the authors employed western blot to identify the expression of HIF-1, GLUT1 and LDHA. They found that HIF-1, GLUT1 and LDHA expression levels in HCT116 cells were decreased after LpEV treatment (Figure 4G & H).

LpEVs inhibited the glycolysis of HCT116 cells

To explore whether LpEVs can inhibit glycolysis in HCT116 cells, the authors detected the changes in glucose consumption, lactic acid production and LDH activity in HCT116 cells after LpEV treatment. LpEVs at a concentration of 200 μg/ml were incubated with HCT116 cells for 48 h and then analyzed using specific detection kits. In HCT116 cells treated with LpEVs, glucose intake (Figure 5A), lactic acid production (Figure 5B) and LDH activity (Figure 5C) were all decreased when compared with the control cells. These findings showed that LpEVs can inhibit glycolysis in HCT116 cells, which alters the viability of colorectal cancer cells.

LpEVs inhibited proliferation & HIF-1α-mediated glycolysis of HCT116 cells in vivo

The authors conducted tumor xenograft tests in nude mice to further investigate whether LpEVs can reduce glycolysis mediated by HIF-1 in vivo and inhibit HIF-1-mediated glycolysis and the development of HCT116 cells. HCT116 and LpEV-treated HCT116 cells were subcutaneously injected into nude mice. After 30 days, the results showed that the tumor sizes in the LpEV treatment group were significantly smaller than those in the control group (Figure 6A). They found that the LpEV-treated group had lower levels of mRNA expression of HIF-1, GLUT1 and LDHA in the excised tumor tissues (Figure 6B–D). The immunohistochemistry examination of tumor sections from all confirmed that HIF-1α, GLUT1 and LDHA expression were reduced in the LpEVs group (Figure 6E). These results in vivo demonstrated that LpEVs exhibited antitumor activities through suppression of HIF-1α, GLUT1 and LDHA in HCT116 cells.

Discussion

Among the men and women who have died of cancer worldwide, the mortality rate of colon cancer is high, which has become a significant burden on human health [22]. In recent years, probiotics as a natural source, which confer anticarcinogenic effects of colon cancer, has been receiving important focus [23]. Secretory metabolites [24], cell-free supernatant and bacterial extracts [25] of Lactobacillus strains have a direct effect on the proliferation, necrosis, apoptosis, migration and invasion of colorectal cancer cells.

Recent studies have found that probiotics can secrete EVs, which is considered a safe and efficient manner to participate in intercellular communication, thus playing a beneficial role in the host [26]. The authors' studies revealed that LpEVs can invade colon cancer cells and prevent their growth. Extracellular vesicles from probiotics were internalized by intestinal epithelial cells through clathrin-mediated endocytosis [27,28], and the entry process was related to temperature [29] and EV size [30]. After being internalized by recipient cells, probiotic-derived EVs play an important role in host physiological and pathological functions. Lactobacillus rhamnosus GG-derived EVs relieved intestinal inflammation and reshaped the gut microbiota [31]. Lactobacillus kefirgranum PRCC-1301-derived EVs inhibited the expression of proinflammatory cytokines in Caco-2 cells and enhanced intestinal barrier function [32]. Lacticaseibacillus paracasei-derived extracellular vesicles induced the expression of endoplasmic reticulum stress-associated proteins to attenuate lipopolysaccharide-induced inflammation in the intestine [17]. Lactobacillus plantarum-derived EVs can overcome chemoresistant colorectal cancer by targeting PDK2 signaling [33]. The authors have previously reported that extracellular vesicles of Lacticaseibacillus paracasei PC-H1 can induce apoptosis of colorectal cancer cells via the PDK1/AKT/BCL-2 pathway [11]. To further explore the different mechanisms of Lacticaseibacillus paracasei PC-H1-derived extracellular vesicles as well as their variable effects at different locations in colon cancer cells, transcriptome sequencing was performed. That research proved that LpEVs alter the expression of glycolysis-related genes and reduce glucose consumption, lactate production and LDH activity in colon cancer cells. Therefore, LpEVs have great prospects in the treatment of cancer. Compared with extracellular vesicles derived from eukaryotic cells or serum, extracellular vesicles derived from bacteria have the advantage of fast reproduction speed and easy acquisition, leading to the large-scale production of extracellular vesicles.

Colon cancer cells exhibit the increase in glycolysis that contributes to their growth and proliferation [34]. HIF-1α, as a transcription factor induced by low-oxygen conditions, can trigger anaerobic glycolysis of tumor cells and promote the proliferation of tumor cells [35]. In addition, HIF-1α can induce overexpression of glycolytic proteins such as the transporter GLUT1 and increase activity of key glycolytic enzymes such as LDHA [36]. Decreased glucose uptake by cancer cells due to the inhibition of glucose transporter GLUT1 [37] and decreased lactic acid production by downregulating LDHA may be the potential therapeutic method in colon cancer diseases [38]. In this study, transcriptome sequencing analysis revealed that the differentially expressed genes were significantly enriched in the HIF-1 signaling pathway. Then the authors demonstrated that LpEVs inhibited colorectal cancer cell glycolysis by suppressing the expression of HIF-1α, GLUT4 and LDHA in vivo and invitro.

The release of LpEVs from parental bacteria involved in bacterial–host cell communication may depend on the variety of genetic materials, proteins and metabolites carried by LpEVs, which act as natural effectors of signaling between bacteria and cells through the transfer of their cargos [39]. Regarding EVs from probiotics, little is known about their cargos. Lactobacillus rhamnosus GG-derived EVs showed protective effects on intestinal barrier through the transfer of their cargo molecules [40]. Lactobacillus casei BL23 extracellular vesicles contain two proteins, P40 and P75, and exert a moderate proinflammatory effect [41]. EVs from Lacticaseibacillus rhamnosus JB-1 contain lipoteichoic acid, which activates Toll-like receptor 2 and an increase of IL-10 production by dendritic cells in an internalization-dependent manner [28]. Even though DNA and RNA have been found in EVs from Lactobacillus casei BL23 and Lactobacillus reuteri BBC3 [42,43], the properties of nucleic acids from probiotics remains to be investigated. sRNAs from Lactobacillus plantarum-derived extracellular vesicles could be transferred into mammalian cells to downregulate TP53 gene, which is connected to apoptosis in HEK293T cells [44]. These findings suggest the importance of EVs in the complex communication network between host and probiotics. In addition, the specific components of LpEVs have not been studied in detail in this paper, and the key ingredient of EVs contents still require more studies to investigate and verify, so as to understand which components play the role of inhibiting colon cancer.

In summary, these findings reveal that LpEVs have the potential to inhibit colorectal cancer growth by decreasing HIF-1α-mediated glycolysis (Figure 7). These findings not only provide a new direction to explore the function mechanism of probiotics but also add a novel perspective for the prevention and treatment of colon cancer.

Conclusion

LpEVs ingested by colorectal cancer can prevent the proliferation of colon cancer cells and inhibit glycolysis of colorectal cancer cells by reducing the expression of HIF-1, GLUT4 and LDHA.